Anticipatory models of load balancing in cloud computing
- Publication Type:
- Issue Date:
Cloud Computing is a recent arrival to the world of IT infrastructure. The concept allows companies to maximise utilisation of their potentials and consequently boost their performance. One of the main benefits of Cloud Computing is the significant increase in efficiency of executing business plans. Additionally, Cloud Computing provides large-scale applications with powerful computing power across global locations. Yet Cloud users are able to share their data easily by using replication methodologies. Cloud Computing structure has been developed based on a multi-tenancy concept. Therefore, availability and efficiency of the resources are important factors in the Cloud architecture. However, as the numbers of users are increasing rapidly, the load will have a significant impact on performance and operation of the Cloud systems. Accordingly, optimised load balancing algorithms that can manage the Cloud load in a time- and cost-efficient manner are required. Much research in recent years has been dedicated to optimising load balancing in Cloud Computing. This optimisation is demonstrated through a balanced network of interacting resources. The goal of this network is to minimise the wait time and maximise utilisation of the throughput. This thesis provides a set of solutions which mitigate the problem of load balancing in the Cloud. The dissertation investigates a novel class of heuristic scheduling algorithms that improves load balancing in workflow scheduling applications. Furthermore, it proposes a new anticipatory replication methodology with the objective of improving data availability to enhance the load balancing between the Cloud sites. In summary, this research innovation implicates the design of optimised load balancing algorithms that consider the magnitude and direction of the load in workflow applications. Furthermore, by architecting the anticipatory replication algorithm, it minimises the numbers of the replicas and enhances the effective network usage in Cloud-based systems.
Please use this identifier to cite or link to this item: